Abstract
Obesity is a significant factor contributing to erectile dysfunction (ED). Early detection of ED generally results in improved treatment outcomes. Our study aims to elucidate the association between relative fat mass (RFM) and ED by analyzing data from the National Health and Nutrition Inspection Survey (NHANES) spanning 2001 to 2004. We used data from NHANES 2001–2004, employing weighted, multivariable-adjusted logistic regression to assess the relationship between RFM and the risk of ED. Additional analyses included subgroup analysis, smooth curve fitting, and threshold effect analysis. Subsequently, the predictive utility of RFM, body mass index (BMI), and waist circumference (WC) for ED was evaluated using the receiver operating characteristic curve and area under the curve (AUC) calculations. A total of 3,947 American male participants were included in this retrospective study within NHANES. Weighted multivariate logistic regression analysis indicated that, after adjusting for potential confounding factors, RFM was positively associated with the risk of ED (OR = 1.03, 95% CI = 1.01-1.05, p < .001). No significant saturation effects between RFM and ED were observed (all ps > .05). In addition, RFM demonstrated superior predictive capability for ED (AUC = 0.644) compared with BMI (AUC = 0.525) and WC (AUC = 0.612). Our findings suggest that higher RFM levels are associated with an increased risk of ED, highlighting its potential utility as a predictive marker for this condition.
Introduction
Erectile dysfunction (ED) is a common clinical condition characterized by a persistent inability to achieve and maintain an erection sufficient for satisfactory sexual performance, affecting men above the age of 20 years (Chen et al., 2024). Beyond traditional causes, such as diabetes and hypertension, lifestyle factors including obesity and inadequate physical activity also play critical roles in predicting ED (Shamloul & Ghanem, 2013). According to the Massachusetts Male Aging study (MMAS), ED is the most prevalent sexual health issue, affecting more than 150 million men globally in 1995, with projections exceeding 300 million by 2025 (Ayta et al., 1999). Evidently, ED significantly detracts from the physical and mental well-being of aging men (Kouidrat et al., 2017). Furthermore, ED has been identified as an independent predictor of cardiovascular diseases and male infertility (Fan et al., 2018; Starc et al., 2019), underscoring the importance of early detection and prevention.
Obesity is defined as abnormal or excessive fat accumulation and has emerged as a significant global public health challenge (Liu et al., 2023). The Obesity Society (TOS) thinks that obesity is intricately linked to numerous common illnesses, including heart disease, cancer, stroke, diabetes, and sexual dysfunction, and may itself constitute a severe debilitating disease (De Lorenzo et al., 2019; Jastreboff et al., 2019). Prior research has established obesity as an independent risk factor for ED, with the incidence of ED correlating with obesity levels (Pizzol et al., 2020). Conversely, numerous clinical studies have demonstrated that weight loss significantly ameliorates ED in obese individuals (Aleid et al., 2017; Kun et al., 2015). Body mass index (BMI) is a widely used metric for obesity. However, BMI’s effectiveness in assessing obesity is limited as it does not differentiate between fat mass, muscle, and bone mass, nor is it specific to gender and age (Piché et al., 2018). In addition, BMI does not accurately reflect abdominal fat distribution. In a cohort study, observations showed relatively weak correlations between BMI and percent abdominal fat mass in adults and older children (Gishti et al., 2015). In response, researchers have developed a new obesity metric, relative fat mass (RFM), which more precisely estimates the percentage of total body fat in adults using a linear equation based on the ratio of height to waist circumference (WC) (Woolcott & Bergman, 2018). Given its reliance on WC measurements, RFM serves as an indicator for both general and abdominal obesity (Woolcott & Seuring, 2023). Numerous studies have shown that RFM is strongly associated with a variety of diseases, including depression (Zhu et al., 2024), coronary heart disease (Efe et al., 2021), periodontitis (Zhao et al., 2024), and type 2 diabetes (Suthahar et al., 2023). Although RFM has been validated as a reliable predictor of various disease risks, its potential relationship with ED has not yet been explored. Thus, our study aims to investigate this relationship through a large-scale observational analysis using data from the National Health and Nutrition Examination Survey (NHANES) from 2001 to 2004.
Method
Study Population From NHANES
Data were extracted from the NHANES database. NHANES is a significant national initiative conducted by the Centers for Disease Control and Prevention’s National Center for Health Statistics (NCHS) to evaluate the health and nutritional status of the U.S. population. NCHS assured informed consent was obtained from all participants. As the study uses de-identified public data, approval from an agency review committee was not required. Because of the lack of data on the presence or absence of ED in other years of NHANES data, we chose datasets from the 2001–2002 and 2003–2004 NHANES cycles for cross-sectional analyses. A previous study showed that RFM values for the 2001–2004 cohort were not significantly different from those for the 2019–2020 cohort. We therefore believe this study is equally predictive of current ED prevalence (Gletsu-Miller et al., 2024). Initially, 21,161 participants were considered, but after applying strict inclusion and exclusion criteria, the final sample comprised 3,947 American adult males from the NHANES 2001–2004 cycle. Of these participants, 3,932 were excluded from the experiment because their lack of height or WC data made it impossible to calculate their RFM. Because ED is a male-specific condition, we excluded all 8,843 female participants. In addition to this, 4,439 participants were also excluded from this experiment due to lack of data related to ED. (Figure 1).
Flowchart of Patient Screening
Assessment of RFM
RFM, a novel computational index developed by American researchers, calculates as follows: for men, 64−(20 × height [cm] / WC [cm]) and for women, 76−(20 × height [cm] / WC [cm]) (Ghulam et al., 2023). The researchers chose to include those index models based on the highest correlation with percent body fat for both women and men, and then derived the formula through linear regression (Woolcott & Bergman, 2018). Anthropometric measurements were performed by trained health technicians at the Mobile Examination Center (MEC), with their performance monitored through direct observation, data reviews, and evaluations by expert inspectors. Data on height and WC are available on the NHANES website. Generally, a higher RFM indicates greater obesity (Kobo et al., 2019). Participants were categorized into four quartiles (Q1–Q4) based on RFM for further analysis. In multiple regression analyses, RFM was treated as both a continuous and a categorical variable, and was considered an exposure factor in our study.
Assessment of ED
Interviews to assess erectile function were conducted in a private room at the MEC, facilitated by audio-computer assistance. The ED self-assessment, used as the outcome variable, was derived from a question in the MMAS (Derby et al., 2000): “Many men experience problems with sexual intercourse. How would you describe your ability to get and maintain an erection adequate for satisfactory intercourse?” Response options included “always or almost always,”“usually,”“sometimes,” and “never.” Based on prior research, ED was defined for participants who answered “sometimes” or “never” when asked about their ability to maintain an erection. Those who responded “always or almost always could” or “usually” were categorized as not having ED (Cao et al., 2023).
Other Covariates Used in NHANES
To mitigate potential confounding effects, demographic characteristics were adjusted in the analysis, including age, race, education, marital status, poverty-income ratio (PIR), smoking status, alcohol consumption, physical activity, hypertension, diabetes, total cholesterol (TC), low-density lipoprotein cholesterol (LDL-C), and triglyceride (TG) levels. Education was categorized into three levels: below high school, high school graduate, and college education or higher. Marital status was classified as either “married or partnered” or “single.” Self-reported medical histories assessed chronic health conditions, determining whether participants had been informed by a health care professional of a specific health issue. Smoking status was defined by having smoked 100 or more cigarettes in one’s lifetime. Alcohol consumption was categorized by whether individuals consumed five or more drinks per day. According to the NHANES database, the definition of physical activity was derived from participants’ self-reported level of leisure time physical activity based on their self-reported level of leisure time physical activity. Based on previous literature (Wang & Ni, 2024), we defined physical activity as performing at least 10 min of moderate activity in the past 30 days.
Statistical Analysis
In the NHANES dataset, multivariate adjusted logistic regression was used to examine the relationship between RFM and ED. Three models were employed: Model 1 (unadjusted), Model 2 (adjusted for age and race only), and Model 3 (adjusted for age, race, education, marital status, PIR, smoking, alcohol consumption, diabetes, hypertension, TC, LDL-C, and TG). Results were reported as odds ratios (ORs) and 95% confidence intervals (CIs). Given NHANES’ complex probability clustering design, weights were incorporated into the statistical analysis. In addition, smooth curve fitting via generalized additive models (GAMs) was employed to assess the nonlinear relationship between RFM and ED. Finally, the predictive efficacy of RFM, BMI, and WC on ED was evaluated using the receiver operating characteristic (ROC) curve and area under curve (AUC) calculations.
Results
Participant Characteristics in NHANES
A total of 3,947 subjects were included in this study, with an average age of 49.55 ± 18.44 years. Among the participants, 1,110 were diagnosed with ED (28.11%). The RFM of ED participants was significantly higher than that of participants without ED. Men with ED were more likely to be older, non-Hispanic White, less educated, married or cohabiting, of lower socioeconomic status, less physically active, smokers, and had higher incidences of hypertension and diabetes. In addition, patients with ED also had higher levels of BMI and TG. Detailed demographic data are presented in Table 1.
Baseline Characteristics of Participants by a History of ED, Weighted
Note. Mean ±SD for continuous variables: the p-value was calculated by the weighted linear regression model. (%) for categorical variables: the p-value was calculated by the weighted chi-square test. RFM = relative fat mass; BMI = body mass index; PIR = ratio of family income to poverty; LDL-C = low-density lipoprotein cholesterol.
Association Between RFM and ED in NHANES
In the NHANES cohort, after adjusting for confounders, a linear correlation was observed between RFM and ED. RFM values were categorized into quartiles, and the positive correlation remained stable across these classifications (trend p < .05). In the highest quartile (Q4), each additional unit increase in RFM was associated with a 39% increase in the risk of ED (Table 2). Smooth curve fitting analysis further confirmed a positive linear relationship between RFM and ED (Figure 2).
The Association Between RFM and ED
Note. Model 1: no covariates were adjusted. Model 2: adjusted for age and race. Model 3: adjusted for age, race, education level, marital status, alcohol drinking, smoking, diabetes, high blood pressure, PIR, total cholesterol, triglycerides, LDL-C. PIR = ratio of family income to poverty; BMI = body mass index; LDL-C = low-density lipoprotein cholesterol; OR = odds ratio; CI = confidence interval; RFM = relative fat mass.
The Nonlinear Association Between the RFM and ED
Subgroup Analyses
Subgroup analyses and interaction tests were conducted to determine if the association between RFM and ED was consistent across the entire population and to identify potential differences among various demographic groups. These analyses were stratified by age, race, marital status, hypertension, and diabetes. Results indicated that the relationship between RFM and ED was independent of these factors (all p-values for interaction > 0.05). The positive correlation between RFM and ED was consistent across different ages, races, and states of diabetes and hypertension, suggesting its applicability to diverse populations (Table 3).
Subgroup Analysis of the Association Between RFM and ED
Note. Age, race, education level, marital status, alcohol drinking, smoking, diabetes, high blood pressure, PIR, total cholesterol, triglycerides, LDL-C were adjusted. PIR = ratio of family income to poverty; BMI = body mass index; LDL-C = low-density lipoprotein cholesterol; OR = odds ratio; CI = confidence interval.
RFM demonstrated a stronger predictive capability for ED than both BMI and WC. Table 4 lists the AUC values for these anthropometric indicators (95% CI): RFM: 0.644 (0.625–0.663), BMI: 0.525 (0.505–0.545), WC: 0.612 (0.592–0.631). Of the three parameters, RFM had the highest AUC. An ROC curve to assess the predictive value of obesity indicators for ED was depicted in Figure 3. In addition, the differences in AUC values between RFM, BMI, and WC were statistically significant (all ps < .001), indicating that RFM may be a more effective predictor of ED than BMI and WC.
The Adiposity Indicators for Predicting ED
Note. AUC = area under curve; 95% CI = 95% confidence interval; BMI = body mass index; CI = confidence interval.
ROC Curve to Assess the Predictive Value of Obesity Indicators for ED Was Depicted
Discussion
The purpose of this study was to explore the association between the RFM index, a newly developed indicator of obesity, and ED in 3,947 men in the United States. Our findings indicate that a higher RFM is associated with more severe ED. Even when RFM is categorized into quartiles (Q1–Q4), the positive correlation persists. Moreover, we employed the ROC curve to evaluate and compare the predictive efficacy of RFM, BMI, and WC for ED. To our knowledge, this is the first study to focus specifically on the correlation between RFM and ED. Currently, BMI is the most commonly used metric for classifying overweight and obesity. Although BMI is straightforward to measure and broadly understood, it has significant limitations in accurately classifying body fat content (Rothman, 2008). BMI does not differentiate between fat and lean mass, which can lead to overestimations in individuals with a normal percentage of body fat and underestimations in others (Paek et al., 2019). With ongoing research, an “obesity paradox” has emerged, wherein BMI does not reliably predict obesity-related outcomes (Reddy et al., 2022; Tom et al., 2018). Some studies have begun to consider WC or visceral obesity as alternatives to BMI (Ross et al., 2020). However, even though WC and other measures of abdominal obesity are deemed potentially superior, they have not consistently demonstrated advantages over BMI in predicting various diseases (Gelber et al., 2008; Huxley et al., 2010). In response to these limitations, the RFM formula was developed to provide a more accurate marker of body fat. Based on the ratio of height to WC, RFM can precisely estimate the proportion of body fat (Caiano et al., 2021). Research by Orison et al. has shown that the body fat percentage measured by RFM aligns with that measured by dual-energy X-ray absorptiometry (DXA) and is consistent across genders and races (Woolcott & Bergman, 2018). Compared with BMI and WC, RFM offers a more accurate description of fat distribution. Furthermore, RFM calculations are gender-specific, providing a more intuitive estimate of body fat percentage for different genders (Suthahar et al., 2022). Our study also found that RFM (AUC = 0.644) is more reliable than BMI (AUC = 0.525) and WC (AUC = 0.612) in predicting ED. RFM requires only height and waist measurements, making it easy to obtain and applicable in both clinical and screening settings. Numerous studies support the association between obesity and ED. An early long-term prospective study suggested that obesity contributes to the development of ED (Bacon et al., 2006), and this has been corroborated by recent observational research (Patoulias et al., 2022). Erection is a complex process involving the regulation of vascular and endocrine factors (Gao et al., 2024). Current studies have confirmed that pathological changes in vascular endothelium can lead to ED (Mitidieri et al., 2020). The adoption of a westernized diet globally has led to prevalent central obesity, characterized by visceral fat accumulation (Higgs, 2023). This type of obesity induces insulin resistance and abnormal levels of circulating adipose-derived factors, contributing to atherosclerosis and reduced endothelium-dependent vasodilation, which directly affects male sexual function, particularly the incidence of ED (Corona et al., 2014). In addition, animal studies have demonstrated that surgical reduction of visceral fat improves endothelial function and enhances erectile function in mice (Zhang et al., 2011). It is important to note that obesity-related changes in sex hormone levels are also critical factors influencing ED. Specifically, a cross-sectional study revealed that obese men with ED have significantly lower testosterone levels (Corona et al., 2008). Beyond the reduction in testosterone, these men also exhibit a decreased ratio of testosterone to estradiol and increased estrogen levels, which significantly influence the physiological processes governing sexual behavior, and play a pivotal role in the development of ED (Boutari et al., 2020). Conversely, a study by Berniza et al. indicated that reproductive hormone levels tend to normalize following bariatric surgery (Calderón et al., 2020). Obesity-related complications such as hypertension also contribute to ED development. Hypertensive patients often experience increased peripheral vascular resistance due to peripheral vascular sclerosis (Johnson et al., 2008). These conditions are mirrored in the corpus cavernosum, resulting in high vascular resistance and inadequate arterial inflow (Diosdado-Figueiredo et al., 2019). A comprehensive survey in the United States reported that approximately 67%–68% of hypertensive men experience some degree of ED (de Oliveira & Nunes, 2021). Physical activity is another significant factor in the development of ED. Research shows that regular exercise improves erectile function, particularly in obese men with a low International Erectile Function Index (IIEF-EF) (Khera et al., 2023). Notably, the most substantial improvements in erectile function were observed with aerobic exercise of moderate intensity or higher (Silva et al., 2017). Thus, we have a reason to speculate that RFM may be positively related to the prevalence of ED.
Our research presents several noteworthy advantages that merit recognition. First, this study used a large, representative sample of the American population to establish the correlation between RFM and ED for the first time, comparing its predictive performance for ED against various obesity metrics. Second, the researchers accounted for potential confounding factors through multiple logical regression analyses and employed subgroup analyses to investigate this correlation across different population demographics. Notably, RFM is easier to measure and more accurately depicts the relationship with ED compared with BMI. Moreover, RFM offers the same measurement precision as DXA without the associated radiation exposure, making it more practical for widespread use.
However, our study is subject to some limitations. The cross-sectional design precludes the determination of a causal relationship between RFM and ED. In addition, recall bias may affect the assessment of ED as it relies primarily on questionnaire responses. Given that the study was conducted solely within the United States, broader surveys are necessary to establish the generalizability of our findings to populations in diverse geographical regions.
Conclusion
In summary, RFM is positively correlated with ED within the American population. RFM serves as a practical obesity index that can be used in the clinical management of ED. Adjusting RFM levels through weight loss may not only delay the progression of ED but also act as a preventive measure. However, higher-quality prospective studies are required to validate our findings.
Footnotes
Acknowledgements
The authors thank all the NHANES who participated in this study, and thank all investigators for making available their data.
Author Contributions
B.Y. and H.W. designed the research. B.Y. and L.T. collected the data. B.Y. and H.W. analyzed the data. B.Y. drafted the manuscript. S.H. and J.F. supervised the study. All authors were involved in writing the case. All authors contributed to the article and approved the submitted version.
Declaration of Conflicting Interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This study was supported by Clinical efficacy evaluation of Dongguan Science and Technology Bureau's high-level special program “Zhixueshengjifang” for the treatment of chronic radiation proctitis and S.H. Dongguan Famous Traditional Chinese Medicine Expert Inheritance Studio.
Ethics Statement
The studies involving human participants were reviewed and approved by the Research Ethics Review Board of the National Center for Health Statistics (NCHS). The patients/participants provided their written informed consent to participate in this study.
Consent for Publication
This manuscript does not include details, images, or videos relating to an individual person; therefore, consent for publication is not required, beyond the informed consent provided by all study participants as described above.
Data Availability
The original contributions presented in the study are included in the article/supplementary material. Further inquiries can be directed to the corresponding author.